Epidemiology of Shigellosis in Lagos, Nigeria: Trends in Antimicrobial Resistance

Transcription

1 J HEALTH POPUL NUTR 2001 Sep;19(3): ICDDR,B: Centre for Health and Population Research ISSN Trends in antimicrobial $ resistance in Nigeria 183 Epidemiology of Shigellosis in Lagos, Nigeria: Trends in Antimicrobial Resistance B.A. Iwalokun 1, G.O. Gbenle 1, S.I. Smith 2, A. Ogunledun 3, K.A. Akinsinde 2, and E.A. Omonigbehin 2 1 Biochemistry Department, College of Medicine, University of Lagos, P.M.B , Lagos; 2 Genetics Division, Nigerian Institute for Medical Research, P.M.B. 2013, Yaba, Lagos; and 3 Medical Microbiology and Histopathology Department, College of Health Sciences, Ogun State University, P.M.B. 2001, Sagamu, Nigeria ABSTRACT Antimicrobial susceptibility of Shigella spp. and Escherichia coli, isolated from diarrhoeal patients in Lagos, was studied from March 1999 to February Four hundred fifty-nine isolates were identified as shigellae (62) and E. coli (397). Shigella flexneri, S. dysenteriae, S. boydii, and S. sonnei accounted, respectively, for 51.6%, 17.7%, 17.7%, and 13% of the total number of shigellae isolated. Eleven cases of shigellosis occurred in the age group of 0 9 years, 22 cases in the age group of years, and 29 cases in the age group of 20 years. Of the 397 E. coli isolates, 11 were enteropathogenic E. coli (EPEC), and 7 of these strains were isolated with shigellae from stools of patients aged 0 9 year(s) (71.4%) and years (28.6%). Over 70% of the Shigella isolates were resistant to two or more drugs, including ampicillin and tetracycline. Twenty-one distinct multidrug resistance patterns were observed in these isolates. During , resistance to ampicillin increased from 70% to 90%, co-trimoxazole from 77% to 85%, chloramphenicol from 71% to 77%, streptomycin from 71% to 79%, and nalidixic acid from 0% to 11.3%. Resistance to tetracycline decreased from 89% to 79% but with MIC 50 values outside the susceptible range. While resistance to ciprofloxacin and ofloxacin remained nil with MIC 50 values of and µg/ml respectively. The results of this study revealed the endemicity of shigellosis with S. flexneri as the predominant serogroup in Lagos. Children and young adults were at a higher risk of severe shigellosis. The results also suggest that ampicillin, tetracycline, co-trimoxazole, and streptomycin should not be used as the first-line drugs in the treatment of shigellosis. Nalidixic acid should still be selectively used for treatment, while ciprofloxacin and ofloxacin can be ideal alternatives. Key words: Dysentery, Bacillary; Shigella; Escherichia coli; Drug resistance, Microbial; Antibiotic resistance; Nigeria INTRODUCTION Shigellosis still remains a public-health problem in most developing countries where communities are ravaged Correspondence and reprint requests should be addressed to: Dr. B.A. Iwalokun Biochemistry Department, College of Medicine University of Lagos P.M.B , Lagos Nigeria bamwal@yahoo.com by poverty, war, poor sanitation, personal hygiene, and water supplies (1). Epidemiologic reports show that about 140 million people suffer from shigellosis with estimated 600,000 deaths per year worldwide (2,3). In Nigeria, shigellosis is still the major cause of dysentery in children aged 0-9 year(s), and many older children are hospitalized almost immediately after the onset of the disease (4). Although antibiotic therapy is important in the treatment of shigellosis in most endemic

2 184 J Health Popul Nutr Sep 2001 Iwalokun BA et al. countries, antibiotics, including ampicillin, trimethoprim-sulphamethoxazole, and nalidixic acid, have been banned in the treatment of shigellosis in countries of Asia and sub-saharan Africa (5,6). Furthermore, drugs, such as fluoroquinolones, azithromycin, and pivamdinocillin, have been found to be efficacious both in vitro and in vivo in the treatment of shigellosis in children and adults (7,8). These drugs have helped control and manage shigellosis in many countries (9). In Nigeria, an active surveillance study on shigellosis was carried out over a decade ago (10). Since then, no studies were conducted to examine the status of drug resistance in shigellosis. This situation might have been responsible for the severity, prolonged illness, and high hospitalization rate of patients with shigellosis in Lagos recently (4). If this trend continues too long, the mortality rate due to shigellosis may rise in Lagos. Furthermore, there is a paucity of information on co-infection of shigellae with other pathogenic Enterobacteriaceae. This study was, therefore, carried out to determine the antibiotic susceptibility profile of Shigella spp. and to determine the incidence of co-infection of shigellae with enteropathogenic Escherichia coli (EPEC). MATERIALS AND METHODS Patients and sample collection During March 1999-February 2000, 1,020 stool samples were collected from diarrhoeal patients who sought treatments at the Massey Children s Hospital, Mainland Infectious Diseases Hospital, Central Medical Health Laboratory Services, Yaba, Lagos, and Randle Medical Health Centre, Lagos. These hospitals/clinics serve as referral centres for most communities and towns in Lagos metropolis. Patients included children, adults, and elderly people. Bacteriology Fresh faeces of patients were collected into Cary-Blair tubes (10 ml per tube) and transported to Microbiology Laboratory of Nigerian Institute of Medical Research, Lagos, for immediate culture. The specimens were inoculated on MacConkey and Salmonella-Shigella agar. Colonies suspected to be shigellae were further subcultured on Simmon-citrate, motility-indole-urea and Kligler-iron agar. The Shigella isolates were speciated biochemically as outlined by Cowan (11). Colonies on MacConkey plates suspected to be E. coli were further tested for enteropathogenicity by slide and tube agglutination tests using E. coli polyvalent antisera A, B, and C (Biotec Laboratories, UK). Antimicrobial susceptibility testing Resistance patterns of the isolated Shigella strains to 12 antibiotics were determined by the agar-diffusion technique (12). Every inoculum was prepared by inoculating 5 ml of Mueller-Hinton broth with five colonies of 18-hour old pure Shigella culture followed by incubation in ambient air and at 37 0 C for 16 hours. The resulting turbid culture was standardized to the turbidity of 0.5 McFarland (A 625nm =0.09) using 0.85% NaCl as the diluent. A sterile cotton swab was dipped into the standardized suspension, drained, and used for inoculating 25 ml of Mueller-Hinton agar in a 100-mm plate (Sterilin, UK). The inoculated plates were air-dried, and antibiotic disks from Oxoid (UK) and Mast Laboratories (Merseyside, UK) were mounted on them. The 12 antibiotics that were tested for susceptibility included: ampicillin, tetracycline, colistin sulphate, cotrimoxazole, cefotaxime, nitrofurantoin, nalidixic acid, streptomycin, ofloxacin, ciprofloxacin, chloramphenicol, and gentamicin. The plates were inverted and incubated in ambient air at 37 C for 18 hours. Zones of inhibition were recorded in millimetres and were compared with those of E. coli ATCC25922, which served as the control. Determination of minimum inhibitory concentration Minimum concentration of each antibiotic inhibitory to the growth of 50% (MIC 50 ) and 90% (MIC 90 ) of the isolates was determined on isosensitivity agar (Oxoid, UK). The agar contained concentration ranges of the antibiotics prepared by two-fold serial dilution according to the National Committee for Clinical Laboratory Standards (13). A multipoint inoculator was used for dispensing 20 ml of adjusted inoculum (10 7 cfu/ml) of each isolate onto the surface of the antibiotic plate to obtain a final inoculum size of cfu/spot. Antibiotic-free plates were inoculated last and were used as negative controls. The positive controls were the plates (one plate per antibiotic tested) inoculated with the reference strain E. coli ATCC MIC 50 of each antimicrobial agent against the Shigella serogroups were evaluated after incubating the plates, containing completely absorbed inocula, in ambient air at 37 0 C for 18 hours (14).

3 Trends in antimicrobial resistance in Nigeria 185 RESULTS Of the 1,020 diarrhoeal stool samples screened, 397 E. coli strains and 62 shigellae strains were identified. While shigellae were isolated from patients ranging in age from 0 to >40 year(s), 35.5% of the patients were aged years. The males outnumbered the females in nearly all age groups (Table 1). (17.7%) cases of shigellosis each (Table 2). Eight cases of shigellosis were caused by S. sonnei. Of the 11 EPEC strains isolated, 7 were isolated with shigellae from diarrhoeal stools of patients aged 0 9 years and year(s) (Table 3). The two age groups were responsible for 71.4% (5 strains) and 28.6% (2 strains) co-infections. Table 1. Age and sex distribution of shigellae and E. coli isolated from diarrhoeal patients E. coli (n=397) Shigella spp. (n=62) Age group (in years) Male % Female % Male % Female % Shigellae were isolated in each month from March 1999 to February 2000 (Fig.). The highest number of isolates was obtained in March, while the lowest numbers were obtained in January and July. Shigella flexneri strains were identified in 32 (51.6%) shigellae-positive cultures, while S. dysenteriae and S. boydii accounted for 11 Resistance to seven antibiotics, such as ampicillin, tetracycline, colistin sulphate, co-trimoxazole, chloramphenicol, streptomycin, and cefotaxime, ranged from 43.5% to 90.3% (Table 4). Resistance to nalidixic acid was 11.3% and to nitrofurantoin and gentamicin 3.2% each. No isolates Fig. Monthly distribution of Shigella isolates obtained from stools of diarrhoeal patients, March 1999-February 2000 Number of Shigella isolates S. dysenteriae S. flexneri S. sonnei S. boydii 1 0 Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Month

4 186 J Health Popul Nutr Sep 2001 Iwalokun BA et al. Table 2. Age and sex distribution of patients with Shigella serogroups isolated Age group Isolates Sex Serogroups (in years) No. % No. of % No. of % S. dysenteriae S. flexneri S. sonnei S. boydii males females No. % No. % No. % No. % Total Table 3. Relative risk of co-infection of shigellae with enteropathogenic E. coli in different age groups Age group Patients with EPEC Patients with Shigellae (in years) With Shigella isolates Without Shigella isolates With EPEC isolates Without EPEC isolates Table 4. Comparison and incidence of resistance to 12 antibiotics among 62 Shigella isolates and serogroups from Lagos, Nigeria Antibiotic % of resistant Resistant Serogroups Shigella isolates isolates S. dysenteriae S. flexneri S. sonnei S. boydii in 1990 (10) No. % No. % No. % No. % No. % Ampicillin Tetracycline Colistin sulphate Co-trimoxazole Chloramphenicol Streptomycin Cefotaxime Nalidixic acid Nitrofurantoin Gentamicin Ofloxacin Ciprofloxacin showed any resistance to ofloxacin and ciprofloxacin. In all, 21 patterns of antibiotic resistance were found among the Shigella serogroups, and further analysis revealed that over 70% of the isolates were resistant to two or more drugs, including ampicillin and tetracycline (Table 5). Minimum inhibitory concentrations of each of the 12 antibiotics that made 50% and 90% of shigellae susceptible in vitro are shown in Table 6. At µg/ ml and µg/ml, ofloxacin and ciprofloxacin inhibited the growth of 50% and 90% of the isolates. Generally, there was a wide variation in MIC 50 and MIC 90 values of the antimicrobial agents, and those of streptomycin, colistin sulphate, tetracycline, chloramphenicol, co-trimoxazole, and ampicillin were very high. DISCUSSION In this study, we isolated 62 Shigella strains from diarrhoeal stools of 1,020 patients, and speciation showed that S. flexneri caused 32 (51.6%) of the total cases of shigellosis. S. dysenteriae and S. boydii were identified in 11 (17.7%) cases each, while eight (13%) cases were caused by S. sonnei. This pattern of shigellosis indicates that S. flexneri is the predominant and most active serogroup in Lagos. Our results agree with the pattern reported by Olukoya and Oni (10). Thus, the predominance of S. flexneri in Lagos has not changed

5 Trends in antimicrobial resistance in Nigeria 187 Table 5. Antibiotic resistance patterns of Shigella isolates Antibiotic resistance pattern No. of resistant strains S. dysenteriae S. flexneri S. sonnei S. boydii (n=11) (n=32) (n=8) (n=11) Amp Tet Amp Tet Cot Amp Tet Str Chl Amp Cot Str Chl Tet Amp Chl Cot Col Tet Amp Cot Col Tet Cef Amp Cot Col Tet Chl Amp Cot Col Str Chl Amp Tet Str Chl Col Str Cot Cef Chl Nal Tet Str Cot Col Chl Amp Cot Col Cef Str Chl Amp Tet Str Cot Col Cef Amp Str Col Cot Cef Chl Amp Cot Col Str Tet Chl Amp Tet Cef Str Nal Nit Col Amp Tet Chl Str Col Cot Nal Cef Amp Tet Col Cot Cef Nal Tet Str Col Cot Chl Gen Amp Tet Col Cot Chl Cef Str Amp Tet Str Col Cot Cef Chl Nit Amp=Ampicillin; Cef=Cefotaxime; Chl=Chloramphenicol; Col=Colistin sulphate; Cot=Co-trimoxazole; Gen=Gentamicin; Nal=Nalidixic acid; Nit=Nitrofurantoin; Str=Streptomycin; Tet=Tetracycline since This is unlike the situation in the islands of Bengal (15) where S. flexneri and S. dysenteriae alternate as most active agents of shigellosis or in endemic communities of Israel and Pakistan where S. sonnei persists as the predominant aetiologic agent (16,17). Table 6. Minimum inhibitory concentrations of each of 12 antimicrobial agents for Shigella isolates (n=62) Antimicrobial agent Range MIC 50 MIC 90 µg/ml µg/ml µg/ml Tetracycline Gentamicin Cefotaxime Co-trimoxazole Ampicillin Nalidixic acid Nitrofurantoin Colistin sulphate Ofloxacin Ciprofloxacin Streptomycin Chloramphenicol Unlike Olukoya and Oni (10), we stratified 62 cases of shigellosis by age and sex, and the pattern obtained shows that both children and young adults were at a higher risk of contacting the disease. This is similar to the distribution pattern found by Khan et al. in Bangladesh (18). The high incidence of shigellosis in young adults might be a reflection of secondary infection caused by contacts with the lower age groups, i.e. 0 4 year(s), 5 9 years, and within 10 to 19-year and 20 to 29-year age groups. Khan et al. reported secondary infection rates of 30.6% and 28.3% in the age groups of 0 4 year(s) and 5 9 years, and secondary infection was observed more in males than females (18). Similarly, the total number of males with shigellosis was more than that of females in the first three groups in our study. Poor personal hygiene and food intake have been documented in these age groups. In a recent study conducted in Lagos, Akinyemi et al. found children (with males dominating) as potential reservoirs of pathogenic Enterobacteriaceae (19). We also examined the incidence of co-infection with EPEC. We found 7 EPEC and Shigella co-infections in

6 188 J Health Popul Nutr Sep 2001 Iwalokun BA et al. the age groups of 0 9-year(s) (71.4%) and years (28.6%). EPEC has been reported as the major cause of acute diarrhoea in children aged 2 years in Latin America (20). Gomez et al. further revealed the significance of EPEC and Shigella co-infections among children who died due to diarrhoea (21). In Nigeria, infant mortality due to diarrhoea was attributed mostly to EPEC in the late 1980s (22). The present trend in Lagos calls for urgent measures to reduce deaths of children due to shigellosis. More so is that shigellosis is endemic in Lagos (10), which is further supported by our results; most of our study children with shigellosis were hospitalized (4). Therefore, public-health strategy should ensure clean water supply, good sewage management, and a clean environment. Furthermore, the results of the antimicrobial susceptibility testing make the management of shigellosis in the population studied complicated. Our antibiograms are at variance with those of Olukoya and Oni (10). When compared with the latter, we found increased resistance rates for ampicillin ( %), streptomycin (71 79%), trimethoprim-sulphamethoxazole (i.e. co-trimoxazole; %), and nalidixic acid (0 11.3%). The decreased resistance rate for tetracycline (89 79%) was not significant for its use, because its MIC 50 values of 64 µg/ml and 128 µg/ml were outside the susceptibility range (13). These antibiotics have been grossly abused in Nigeria (23,24). Since there are no antibiotics with poor in-vitro efficacy against shigellae that have good clinical cure rate, our results suggest that these drugs, except nalidixic acid, should not be used in the treatment of shigellosis in any age groups in Lagos. Nalidixic acid was exempted, because it demonstrated over 78% bacteriologic cure (13) with MIC 50 values within the susceptibility range (Table 6). However, the upsurge of resistance to nalidixic acid by shigellae, as observed in our study, mimics trends in communities where nalidixic acid was introduced to cure shigellosis caused by ampicillin and trimethoprimsulphamethaxazole-resistant strains (7). Henceforth, to prevent a situation where the non-use of nalidixic acid in the treatment of shigellosis will be considered, it is imperative that the drug should be used in patients, especially in children, whose aetiologic agents are susceptible in vitro. Nalidixic acid remains the drug of choice for shigellosis in Pakistan (17). The drug maintains appreciable intracellular concentrations that are ideal for efficacy against inflammatory diarrhoeal illnesses, such as shigellosis (25). Our results further confirm the in-vitro bacteriologic efficacy of gentamicin and nitrofurantoin as reported previously (10). Since these drugs possess poor intracellular concentrations, these are ineffective for a clinical cure (26). Unlike nalidixic acid, ciprofloxacin and ofloxacin retain their 100% efficacies that they possessed 10 years ago against bacteriologic shigellae. They are, thus, excellent drugs either for first-line treatment or for backup treatment of shigellosis in Lagos. However, the drugs are restricted for use in children due to their reported toxicity (27) and are relatively expensive when used in adults. In Lagos, the cost of ciprofloxacin or ofloxacin ranges from $5.00 to $7.00 for a five-day therapy in adults. Since nalidixic acid is apparently the only available effective drug in use in the treatment of shigellosis in Lagos, there is a need to employ other antibiotics with proven bacteriologic and clinical efficacy against shigellosis in other endemic communities. Such drugs include macrolides, such as azithromycin (7), pivmecillinam (8), and third-generation cephalosporins, such as cefixime and ceftazidime (9). Our result was inconclusive to suggest the non-use of cefotaxime as the first-line drug, although the resistance rate of 43.5% was obtained. More than 62% of the S. sonnei and S. boydii strains were still susceptible to cefotaxime, suggesting that this drug could still be clinically relevant when used selectively and as required in the treatment of shigellosis in Nigeria. We conclude that government and non-government organizations have a significant role to play in ensuring the procurement and distribution of new shigellocidal drugs at affordable prices, providing funds and machinery for continuous surveillance, and making longlasting policies that will solve the problems of illegal drug trade, inappropriate drug prescription, and selfmedication in Nigeria. On a global scale, the current tempo toward the production of live oral and parenteral subunit vaccines against shigellosis should be raised. Vaccines, such as SC602 (Shigella flexneri 2a), which have been subjected to phase-1 clinical trials and found safe, immunogenic, and protective should be considered for further trials, and their usage be legalized in most endemic communities (28). It is only then that the publichealth burden of shigellosis can be curtailed drastically. ACKNOWLEDGEMENTS The authors are grateful to Ms Ibiyade Khafayet for typing the mannuscript.

7 Trends in antimicrobial resistance in Nigeria 189 REFERENCES 1. Shears P. Shigella infections. Ann Trop Med Parasitol 1996;90: World Health Organization. Vaccine research and development: new strategies for accelerating Shigella vaccine development. Wkly Epidemiol Rec 1997;72: World Health Organization. Diarrhoeal disease due to Shigella disease. In: Vaccines, immunization and biologicals. Geneva: World Health Organization, 1998: Niemogha MT, Alabe SA, Uzoma KL, Odugbemi TO, Adegbola RA, Coker AD. The incidence of Salmonella, Shigella and other enteric bacterial pathogens in stool specimens of diarrhoea patients. Niger Med J 1995;28: Bennish ML, Salam MA, Hossain MA, Myaux J, Khan EH, Chakraborty J et al. Antimicrobial resistance to Shigella isolates in Bangladesh, : increasing frequency of strains multiply resistant to ampicillin, trimethoprim sulfamethoxazole, and nalidxic acid. Clin Infect Dis 1992;14: Bogaerts J, Verhaegen J, Munyabikali JP, Mukantabana B, Lemmens P, Vandeven J et al. Antimicrobial resistance and serotypes of Shigella isolates in Kigali, Rwanda (1983 to 1993): increasing frequency of multiple resistance. Diagn Microbiol Infect Dis 1997;28: Khan WA, Seas C, Dhar U, Salam MA, Bennish ML. Treatment of shigellosis: V. Comparison of azithromycin and ciprofloxacin. A double blind randomized, controlled trial. Ann Intern Med 1997;126: Kabir I, Rahaman MM, Ahmed SM, Akhter SQ, Butler T. Comparative efficacies of pivmecillinam and ampicillin in acute shigellosis. Antimicrob Agent Chemother 1984;25: Moolasart P, Eampokalap B, Ratanasrithong M. Comparison of the efficacy of ceftibuten and norfloxacin in the treatment of acute gastrointestinal infection in children. Southeast Asian J Trop Med Public Health 1999;30: Olukoya DK, Oni O. Plasmid profile analysis and antimicrobial susceptibility patterns of Shigella isolates from Nigeria. Epidemiol Infect 1990;105: Cowan ST. Manual for identification of medical bacteria. 2d ed. London: Cambridge University Press, 1974: Bauer AW, Kirby WMM, Sherris JC, Turck M. Antibiotic susceptibility testing by a standard single disk method. Am J Clin Pathol 1966;36: National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testiing: sixth international supplement. Wayne, PA: National Committee for Clinical Laboratory Standards, (NCCLS document no. MI00-S6). 14. Phillips I, Andrews JM, Bridson E, Cooke EM, Holt HA, Spencer RC et al. A guide to sensitivity testing. Report of the Working Party on Antibiotic Sensitivity Testing of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother 1991; 27(Suppl D): Ghosh AR, Sehgal SC. Shigella infections among children in Andaman -an archipelago of tropical islands in Bay of Bengal. Epidemiol Infect 1998;121: Mates A, Eyny D, Philo S. Antimicrobial resistance trends in Shigella serogroups isolated in Israel, Eur J Clin Microbiol Infect Dis 2000;19: Mahmood A. Nalidixic acid is still the drug of choice for shigellosis in Pakistan. J Pak Med Assoc 2001;51: Khan M, Curlin GT, Huq l. Epidemiology of Shigella dysenteriae type l infections, in Dacca urban area. Trop Geogr Med 1979;31: Akinyemi KL, Olukoya DK Oyefolu AO, Amorighoye EP, Omonigbehin EO. Prevalence of multiple drug resistance Salmonella typhi among clinically diagnosed typhoid fever patients in Lagos, Nigeria. Z Naturforsch 2000;55: Notario R, Borda N, Gambande T, Sutich E. Species and serovars of enteropathogenic agents associated with acute diarrheal disease in Rosario, Argentina. Rev Inst Med Trop Sao Paulo 1996;38: Gomes TAT, Rassi V, MacDonald KL, Ramos SRTS, Trabulsi LR, Vieira MAM et al. Enteropathogens associated with acute diarrheal disease in urban infants in São Paulo, Brazil. J Infect Dis 1991;164: Bojuwoye BJ. Pathogenic mechanisms of bacterial diarrhoea: new insight and clinical implications. Nig Med Parasitol 1988;82:185-8.

8 190 J Health Popul Nutr Sep 2001 Iwalokun BA et al. 23. Olukoya DK, Daini O, Alabi SA, Coker AO, Odugbemi T, Akinrinmisi EO. Antimicrobial resistance patterns and plasmids of enteropathogenic Escherichia coli isolated in Nigeria. Eur J Epidemiol 1988;4: Alabi AA, Tolu O. Antimicrobial susceptibility pattern of Aeromonas and Plesiomonas strains isolated from patients with diarrhoea in Nigeria. Cent Afr J Med 1990;36: De Mol P, Mets T, Lagasse R, Vandepitte J, Mutwewingabo A, Butzler J-P. Treatment of bacillary dysentery: a comparison between enoxacin and nalidixic acid. J Antimicrob Chemother 1987;19: Haltalin KC, Nelson JD. Failure of furazolidone therapy in shigellosis. Am J Dis child 1972;123: Schaad UB, Salam MA, Aujard Y, Dagan R, Green SDR, Peltola H et al. Use of fluoroquinolones in pediatrics: consensus report of an International Society of Chemotherapy Commission. Pediatr Infect Dis J 1995;14: Coster TS, Hoge CW, Van De Verg LL, Hartman AB, Oaks EV, Venkatesan MM et al. Vaccination against shigellosis with attenuated Shigella flexneri 2a strain SC602. Infect Immun 1999;67: [AUTHOR: 1. Citation no. 28 not cited in the text?